This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Evaluate the Safety, Efficacy and Dose Response of GSK573719 in Combination With Fluticasone Furoate in Subjects With Asthma (NCT01573624), a randomized phase 2 study. The primary outcome is Model Predicted Change From Baseline Trough Force Expiratory Volume in 1 Second (FEV1).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01573624 |
| Phase | Phase 2 |
| Condition | Asthma |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Double |
| Enrollment | 421 |
| Arms in the primary outcome | 5 |
| Lead sponsor | GlaxoSmithKline |
| Results posted | Yes |
2. Clinical Question
The central question was whether adding different doses of GSK573719 to fluticasone furoate was associated with a dose-related change in lung function in people with asthma. The registry describes a randomized, double-blind crossover study and identifies trough FEV1 change from baseline as the primary endpoint. Fluticasone furoate alone and fluticasone furoate with vilanterol were included as active comparators.
Population
People with asthma who were eligible for the trial. The registry reports enrollment of 421 participants and defines the primary analysis population as randomized participants who received at least one dose of study medication, with analyses limited to those with data at the relevant time points.
Intervention
Fluticasone furoate 100 mcg combined with one of five listed UMEC doses: 15.6 mcg, 31.25 mcg, 62.5 mcg, 125 mcg, or 250 mcg. The trial description characterizes the study as evaluating GSK573719 in combination with fluticasone furoate.
Comparator
Fluticasone furoate 100 mcg alone was the listed primary comparison reference. Fluticasone furoate with vilanterol was also included as an active control.
Primary question
What change from baseline in trough FEV1 was predicted for each fluticasone furoate and UMEC dose combination, with the expected response adjusted by fluticasone furoate 100 mcg alone?
This question concerns estimated lung-function changes over treatment periods, not long-term asthma outcomes. The registry’s secondary measures include morning and evening peak expiratory flow and rescue albuterol or salbutamol use. The results therefore address several short-term respiratory measures, but the primary endpoint is the modeled trough FEV1 change.
3. Trial Design
The registry describes NCT01573624 as a completed Phase 2, randomized, double-blind, multicenter crossover study. Enrollment was 421 participants. The listed start date is 2012-04-03, and the primary completion date is 2013-02-04. The registry identifies asthma as the condition under study and GlaxoSmithKline as the sponsor.
The planned treatment set comprised seven potential treatments: five fluticasone furoate and GSK573719 dose combinations, fluticasone furoate alone, and fluticasone furoate with vilanterol. Each eligible participant was to receive a sequence of three of those treatments across three treatment periods. The detailed description says that the inhaled GSK573719 treatments were given over 14 days and that total participation was approximately 14 weeks. Period-specific baseline and treatment measurements were used for the primary endpoint.
A crossover design means participants receive more than one treatment over time, rather than remaining in only one treatment group. In general, this permits comparisons that account for repeated observations within participants. The registry record does not report enough detail to establish how sequence, period, or carryover effects were handled in the analysis, so those features should not be inferred from the design label alone.
The primary endpoint analysis used the registry-defined Intent-to-Treat population: participants randomized to treatment who received at least one dose. The endpoint description further says that only participants with data at the indicated time points were analyzed. The record does not report the number of participants contributing to each primary estimate, the detailed allocation of treatment sequences, or the reasons for missing observations. The active comparator dose is listed in the results as fluticasone furoate 100 mcg plus vilanterol 25 mcg; the registry’s study description also refers to the combination as 100/25 mcg.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Before assignment: A total of 706 participants were screened for this study designed in 3 period crossover, incomplete block manner; 523 participants entered 2-week open label run-in period where they received fluticasone furoate (FF) 100 micrograms (mcg) dry powder inhalation once daily. A total of 421 participants were randomized to double-blind treatment period.
Only the first 6 of 7 groups are shown.
Period 1 : Treatment Period 1 (14 Days)
| Stage | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| STARTED | 64 | 62 | 60 | 63 | 58 | 55 |
| COMPLETED | 58 | 59 | 51 | 61 | 56 | 51 |
| NOT COMPLETED | 6 | 3 | 9 | 2 | 2 | 4 |
Reasons for not completing this period:
| Reason | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| Adverse Event | 0 | 0 | 1 | 0 | 0 | 1 |
| Lack of Efficacy | 4 | 3 | 6 | 1 | 2 | 0 |
| Protocol Violation | 1 | 0 | 0 | 0 | 0 | 0 |
| Withdrawal by Subject | 1 | 0 | 1 | 0 | 0 | 2 |
| Protocol-defined stopping criteria | 0 | 0 | 1 | 1 | 0 | 1 |
Period 2 : Washout Period 1 (12-14 Days)
| Stage | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| STARTED | 58 | 59 | 51 | 61 | 56 | 51 |
| COMPLETED | 57 | 58 | 48 | 56 | 56 | 48 |
| NOT COMPLETED | 1 | 1 | 3 | 5 | 0 | 3 |
Reasons for not completing this period:
| Reason | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| Adverse Event | 0 | 0 | 0 | 0 | 0 | 1 |
| Lack of Efficacy | 0 | 1 | 0 | 1 | 0 | 0 |
| Protocol Violation | 1 | 0 | 2 | 1 | 0 | 0 |
| Physician Decision | 0 | 0 | 0 | 0 | 0 | 1 |
| Withdrawal by Subject | 0 | 0 | 0 | 2 | 0 | 0 |
| Protocol-defined stopping criteria | 0 | 0 | 1 | 1 | 0 | 1 |
Period 3 : Treatment Period 2 (14 Days)
| Stage | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| STARTED | 57 | 58 | 48 | 56 | 56 | 48 |
| COMPLETED | 50 | 52 | 45 | 55 | 53 | 43 |
| NOT COMPLETED | 7 | 6 | 3 | 1 | 3 | 5 |
Reasons for not completing this period:
| Reason | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| Lack of Efficacy | 5 | 4 | 2 | 1 | 3 | 2 |
| Protocol Violation | 0 | 1 | 0 | 0 | 0 | 2 |
| Physician Decision | 1 | 0 | 0 | 0 | 0 | 0 |
| Withdrawal by Subject | 0 | 1 | 1 | 0 | 0 | 1 |
| Protocol-defined stopping criteria | 1 | 0 | 0 | 0 | 0 | 0 |
Period 4 : Washout Period 2 (12-14 Days)
| Stage | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| STARTED | 50 | 52 | 45 | 55 | 53 | 43 |
| COMPLETED | 49 | 48 | 43 | 53 | 52 | 41 |
| NOT COMPLETED | 1 | 4 | 2 | 2 | 1 | 2 |
Reasons for not completing this period:
| Reason | FF 100 mcg First | FF 100 mcg + UMEC 15.6 mcg First | FF 100 mcg + UMEC 31.25 mcg First | FF 100 mcg + UMEC 62.5 mcg First | FF 100 mcg + UMEC 125 mcg First | FF 100 mcg + UMEC 250 mcg First |
|---|---|---|---|---|---|---|
| Adverse Event | 1 | 0 | 0 | 0 | 0 | 0 |
| Lack of Efficacy | 0 | 1 | 1 | 1 | 1 | 0 |
| Protocol Violation | 0 | 1 | 0 | 0 | 0 | 0 |
| Lost to Follow-up | 0 | 1 | 0 | 0 | 0 | 0 |
| Withdrawal by Subject | 0 | 0 | 0 | 0 | 0 | 1 |
| Protocol-defined stopping criteria | 0 | 1 | 1 | 1 | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: The primary endpoint was analyzed using Intent -to -Treat (ITT) Population defined as all participants randomized to treatment and who received at least one dose of study medication. Only those participants with data available at the indicated time points were analyzed.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | All Treatments Combined |
|---|---|---|
| Participants analysed | 421 | |
| Age, Continuous | Years | 47.5 (SD 13.84) |
| Sex: Female, Male: Female | Participants | 289 |
| Sex: Female, Male: Male | Participants | 132 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 36 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 18 |
| Race (NIH/OMB): White | Participants | 367 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Model Predicted Change From Baseline Trough Force Expiratory Volume in 1 Second (FEV1), measured over Baseline (Day 1) and Day 15 of each treatment period.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Litres (L) |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was model-predicted change from baseline in trough forced expiratory volume in one second, or FEV1, measured in litres. FEV1 is defined in the registry as the maximal amount of air that can be forcefully exhaled in one second. It is used as a measure of lung function and provides a continuous outcome that can represent changes in respiratory function over a treatment period.
For each participant and treatment period, baseline was the pre-dose FEV1 on Day 1. The endpoint description defines the trough value as the FEV1 obtained 24 hours after morning dosing on Day 14. Change from baseline was the difference between that on-treatment value and the period’s baseline value. The highest FEV1 from three acceptable spirometric efforts was recorded within the stated morning measurement window after withholding albuterol or salbutamol for at least 4 hours.
The registry reports that a slope-intercept on log dose model was used to predict trough FEV1 change for the fluticasone furoate and UMEC combinations, adjusted by fluticasone furoate 100 mcg alone. The estimates are means with associated 95% confidence intervals. This modeled endpoint is intended to characterize the dose response, rather than simply describe an unadjusted change in an individual treatment group. The registry record does not report a threshold for a clinically important FEV1 change, so statistical estimates alone do not establish the practical importance of any observed difference.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary estimates rise across the listed dose levels: predicted change from baseline was 0.026 L for fluticasone furoate 100 mcg plus UMEC 15.6 mcg, 0.032 L for 31.25 mcg, 0.038 L for 62.5 mcg, 0.044 L for 125 mcg, and 0.051 L for 250 mcg. These are model-predicted mean changes, not raw individual observations. The pattern is consistent with increasing predicted change across the dose series, but it should be read as a description of the fitted estimates rather than proof of a particular biological mechanism.
The 95% confidence intervals communicate uncertainty around each estimate. For the 15.6 mcg, 31.25 mcg, and 62.5 mcg estimates, the reported intervals include zero. For the 125 mcg and 250 mcg estimates, the reported intervals are above zero. The intervals still have appreciable width, and overlap between intervals is not, by itself, a formal test of differences between dose estimates. No dose-by-dose inferential tests or multiplicity adjustment for the primary dose-response estimates are reported in the registry record.
The active comparator and fluticasone furoate-alone treatment provide context for the study, but the primary description specifically presents modeled expected responses for the combination doses adjusted by fluticasone furoate alone. The registry does not provide primary endpoint estimates for the vilanterol active-control arm in the posted primary results. It would therefore be inappropriate to draw a direct comparative conclusion about that control from these estimates.
The endpoint was measured at the end of a short treatment period, with the primary timeframe defined around baseline and Day 15, while the endpoint description identifies the trough measurement as Day 14 after morning dosing. This is evidence about a short-term lung-function measure under the recorded study conditions. It does not establish the persistence of an effect, changes in symptoms or exacerbations, or the balance of benefits and harms over longer use. Those broader questions are not answered by the primary estimates alone.
Enrollment was 421, but the number contributing to each primary estimate differs by dose, ranging from 146 to 155 in the posted results. In a crossover study, those figures should not be interpreted as mutually exclusive sets of participants assigned permanently to separate parallel arms. The registry’s stated analysis population and time-point availability also mean that the estimates reflect participants with relevant observed data. Missing-data methods beyond that description are not reported in the registry record.
9. Statistical Methods Explained
The primary analysis is described as a dose-response model for a continuous outcome. FEV1 is measured in litres, and the reported estimate is the model-predicted mean change from baseline in trough FEV1. The registry specifies a slope-intercept on log dose model and says the predicted response was adjusted by fluticasone furoate 100 mcg alone. This makes the estimates model-based summaries of the dose pattern, not simple averages that can be interpreted without reference to the model.
What a mean estimates
A mean is a measure of the centre of a set of continuous observations. In this trial, the primary results are model-predicted means of change from baseline. A positive change indicates that the measured trough FEV1 value was higher than the period-specific baseline on the scale of the endpoint; a negative change would indicate a lower value. The mean does not show the range of individual participant responses or how many participants had changes in either direction. The registry’s primary result reports confidence intervals rather than a measure of individual variability.
How to read a confidence interval
A confidence interval expresses the uncertainty around an estimated value under the analysis model. The 95% intervals around the primary estimates indicate a range of values compatible with the data and model at that confidence level. A narrower interval would convey greater precision than a wider interval, but precision does not determine whether a difference is important in practice. Whether an interval includes zero is relevant to whether the data distinguish the estimated change from no change under conventional statistical interpretation; it does not, by itself, provide a complete account of dose comparisons, multiplicity, or clinical relevance.
The primary estimates have a continuous scale and show a gradual increase across the five listed combination doses. A fitted dose-response curve can summarize a pattern across doses, but its interpretation depends on the form of the model and the observations supporting it. The registry identifies the model form but does not report the coefficient estimates, model diagnostics, goodness-of-fit information, or a formal test for a dose trend. Those details are not reported in the registry record.
Why the crossover design matters
In general, a crossover study measures participants under multiple treatment conditions at different times. This can make within-participant information useful because each participant may contribute observations under more than one treatment. It also makes period and treatment sequence relevant: measurements can vary over time, and effects from an earlier treatment may potentially influence a later period. The registry confirms three treatment periods and a sequence of three selected treatments per participant, but it does not report the analysis specification for treatment sequence, period effects, or carryover. Those aspects should not be assumed to have been controlled in a particular way.
The primary population is defined as randomized participants who received at least one dose. The endpoint description adds that only participants with data at the indicated time points were included in the analysis. This is a stated population rule, not a complete missing-data strategy. The registry record does not report how missing measurements were handled, whether imputation was used, or how sensitivity to missingness was assessed.
Reported analyses and recalculations
For secondary outcomes, the record includes sponsor-reported mixed-model analyses as well as simple unadjusted recalculations from posted summary statistics. A recalculated comparison is an approximation based on those summaries and does not reproduce the sponsor’s full model, which may use the crossover structure and other analysis details. When considering a discrepancy between a recalculation and the sponsor’s own analysis, the sponsor’s own analysis takes precedence. The posted record does not identify a discrepancy flag for this trial, but the distinction in method remains important: the two types of estimate do not necessarily represent the same statistical procedure.
For example, the morning PEF sponsor-reported net mean differences for the two posted comparisons are 15.9 and 17.4 L/min, each with p < 0.001. The corresponding recalculated confidence intervals are 9.137 to 22.66 and 10.61 to 24.19. These are not interchangeable outputs: the former are explicitly reported as mixed-model results, while the recalculations use posted summaries and approximations. The sponsor-reported evening PEF estimates and rescue-use analyses likewise have their own reported confidence intervals and p-values.
Planning illustrations are not observed power
Sample-size illustrations based on observed effects answer a hypothetical planning question: how many participants per arm might be needed under stated assumptions to detect a chosen fraction of an observed effect at a specified power level. The registry’s illustration assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such calculations are not the trial’s design justification, are not the observed power of the completed trial, and do not show that a future study would reproduce the same effect. The trial’s own sample-size justification is not reported in the registry record.
For continuous outcomes, over-dispersion means that observed variability is greater than a statistical model assumes. In general, if variability is underestimated, uncertainty intervals can be too narrow and tests can appear more conclusive than warranted. The posted recalculations use approximate variability inferred from summary standard errors; they are not a substitute for individual-level model fitting. The registry record does not provide the raw data needed to independently assess the fitted models or their assumptions.
10. Multiplicity, Missing Data and Analysis Populations
The registry identifies a primary endpoint and lists six outcome measures, with three secondary outcomes posted. It does not report a multiplicity strategy for the set of secondary endpoints, dose comparisons, or analyses. It is therefore unclear from the registry record whether formal adjustment for multiple testing was planned or performed. Unadjusted p-values should be read as tests for individual comparisons, not as control of the overall false-positive risk across all outcomes and comparisons.
The primary endpoint is a modeled dose-response outcome, but the registry does not report a formal primary hypothesis-testing procedure or a multiplicity plan for interpreting the several dose estimates. Confidence intervals for individual dose estimates describe their uncertainty; they do not automatically answer whether differences between doses are statistically distinguishable. The record also does not report a prespecified hierarchy for interpreting secondary outcomes or the active-control comparison.
The primary population is described as all randomized participants who received at least one dose of study medication, with analysis limited to people who had data at the indicated time points. This is the registry’s stated analysis population. The record does not report whether all randomized participants were included in every secondary analysis, the amount or pattern of missing data, or whether missingness was related to treatment or outcome. It also does not report imputation procedures or sensitivity analyses for missing observations. Accordingly, the effect of missing data on the estimates cannot be assessed from the posted description alone.
In general, missing outcome data can reduce precision and can introduce bias if the participants with observed measurements differ systematically from those without measurements. The direction and size of any such influence depend on the missingness pattern and analysis assumptions; neither should be guessed without evidence. For this crossover study, the record gives the number contributing to each listed estimate, but not a complete account of period-specific missingness or reasons for withdrawal.
The registry’s use of an Intent-to-Treat label should be read alongside its stated requirement for at least one dose and available data at the relevant time points. The record does not report additional details that would clarify whether a stricter all-randomized-participant analysis was conducted. A fuller statistical analysis plan, missing-data description, and multiplicity procedure are not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| FF 100 mcg | 0 | 187 | 0.0% |
| FF 100 mcg + UMEC 15.6 mcg | 0 | 183 | 0.0% |
| FF 100 mcg + UMEC 31.25 mcg | 1 | 179 | 0.6% |
| FF 100 mcg + UMEC 62.5 mcg | 0 | 180 | 0.0% |
| FF 100 mcg + UMEC 125 mcg | 0 | 176 | 0.0% |
| FF 100 mcg + UMEC 250 mcg | 0 | 186 | 0.0% |
| FF 100 mcg + VI 25 mcg | 1 | 172 | 0.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| FF 100 mcg | 0 of 187 |
| FF 100 mcg + UMEC 15.6 mcg | 0 of 183 |
| FF 100 mcg + UMEC 31.25 mcg | 0 of 179 |
| FF 100 mcg + UMEC 62.5 mcg | 0 of 180 |
| FF 100 mcg + UMEC 125 mcg | 0 of 176 |
| FF 100 mcg + UMEC 250 mcg | 0 of 186 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 3% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| FF 100 mcg | 3 of 187 |
| FF 100 mcg + UMEC 15.6 mcg | 4 of 183 |
| FF 100 mcg + UMEC 31.25 mcg | 2 of 179 |
| FF 100 mcg + UMEC 62.5 mcg | 5 of 180 |
| FF 100 mcg + UMEC 125 mcg | 4 of 176 |
| FF 100 mcg + UMEC 250 mcg | 7 of 186 |
Most frequent serious adverse event terms
| Event term | Organ system | FF 100 mcg | FF 100 mcg + UMEC 15.6 mcg | FF 100 mcg + UMEC 31.25 mcg | FF 100 mcg + UMEC 62.5 mcg | FF 100 mcg + UMEC 125 mcg | FF 100 mcg + UMEC 250 mcg |
|---|---|---|---|---|---|---|---|
| Asthma | Respiratory, thoracic and mediastinal disorders | 0 of 187 | 0 of 183 | 1 of 179 | 0 of 180 | 0 of 176 | 0 of 186 |
| Cholecystitis acute | Hepatobiliary disorders | 0 of 187 | 0 of 183 | 0 of 179 | 0 of 180 | 0 of 176 | 0 of 186 |
Most frequent non-serious adverse event terms
| Event term | Organ system | FF 100 mcg | FF 100 mcg + UMEC 15.6 mcg | FF 100 mcg + UMEC 31.25 mcg | FF 100 mcg + UMEC 62.5 mcg | FF 100 mcg + UMEC 125 mcg | FF 100 mcg + UMEC 250 mcg |
|---|---|---|---|---|---|---|---|
| Headache | Nervous system disorders | 3 of 187 | 4 of 183 | 2 of 179 | 5 of 180 | 4 of 176 | 7 of 186 |
Serious events with the highest arm-level rate: cholecystitis acute (0.6%), asthma (0.6%).
The posted safety figures list participants at risk, participants affected by serious adverse events, participants affected by other adverse events, and deaths for each treatment group. Serious adverse events are uncommon in these listings: the fluticasone furoate plus UMEC 31.25 mcg group and the fluticasone furoate plus vilanterol 25 mcg group each have one participant listed with a serious adverse event; the other listed groups have zero. Deaths are listed as zero in every group.
The registry lists acute cholecystitis and asthma among the leading serious adverse-event terms, each at 0.6%. The record does not provide a term-by-treatment mapping in the registry record, so those terms should not be assigned to a specific group. The other adverse-event counts range from 2 to 7 participants across the listed groups; these are counts, not rates adjusted for exposure time.
These figures describe events reported during the trial and should not be interpreted as proof that a treatment is safe or unsafe. Group sizes differ, participants in a crossover study can contribute exposure to more than one treatment, and the registry record does not report detailed event timing, duration, severity classifications beyond serious versus other, or exposure-adjusted analyses. The counts also do not establish whether an event was caused by treatment. Safety interpretation is limited to the information recorded.
12. What the Result Does, and Does Not, Mean
The primary results describe a modeled pattern in short-term trough FEV1 change across five fluticasone furoate and UMEC dose combinations. The predicted mean changes increase across the listed dose levels, and the confidence intervals indicate differing degrees of uncertainty around those estimates. This is a statistical description of the posted outcome, not a treatment recommendation or a conclusion about long-term asthma control.
The study’s crossover design, short treatment periods, and specific endpoint define the scope of the evidence. The results concern participants with asthma who met the trial’s eligibility criteria and had relevant data available, under the recorded dosing and measurement conditions. The registry record does not establish whether the findings apply to people outside that study population or to other patterns of use.
A confidence interval that includes zero does not show that the true effect is exactly zero; it indicates that the estimate is not precise enough to rule out zero under the stated interval. Likewise, an interval that lies above zero does not establish that the magnitude is clinically important. The registry does not report a threshold for clinical importance, and the reported estimates do not describe individual responses.
The secondary outcomes add context but differ from the primary endpoint and include multiple comparisons without a reported multiplicity procedure. Their unadjusted results should therefore not be treated as independent confirmation. Safety counts are also descriptive, with no deaths listed, but small event counts and incomplete detail limit what can be inferred about comparative safety.
The registry record does not report enough detail to assess all model assumptions, missing-data sensitivity, treatment-period effects, carryover, or the prespecified statistical testing framework. These limitations do not erase the posted estimates; they define the boundaries of what the estimates can support. In particular, the results do not establish a long-term effect, a clinical benefit threshold, or a favorable overall balance of outcomes.
13. Why This Trial Matters Statistically
This trial is statistically notable because it combines a crossover design with a dose-ranging primary question. That structure produces a series of modeled dose-specific estimates rather than a simple set of independent parallel-group means. Understanding the distinction matters when interpreting both the precision of the primary estimates and the secondary comparisons.
The primary endpoint is a continuous lung-function measure, and the record reports means with 95% confidence intervals. This allows readers to consider both the estimated direction and the uncertainty, rather than relying on a single point estimate. The dose pattern is visible in the sequence of predicted values, while the intervals show that precision varies across estimates. A pattern in point estimates is not, by itself, a formal test that every adjacent dose differs.
The record also illustrates why analysis labels matter. Sponsor-reported mixed-model results for selected secondary comparisons are distinct from simple unadjusted recalculations based on summary statistics. The latter are useful as approximate checks or planning inputs, but they do not replace the trial’s own model-based analysis. Keeping those outputs separate helps prevent false precision or accidental overstatement.
Finally, interpretation depends on what the registry does not explain. The record identifies the analysis population but gives limited information about missing-data methods, multiplicity control, model diagnostics, and crossover-specific effects. These omissions make it harder to judge how robust the estimates are to analytical choices. The resulting statistical lesson is not that the results should be discarded, but that estimates, uncertainty, design, and reporting completeness all need to be considered together.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The registry record leaves several trial-specific questions unresolved. Although the design is described as a randomized crossover with three treatment periods, the sequence allocation and analysis of treatment, period, and possible carryover effects are not reported. Those details matter because each participant could receive multiple treatments over time, and measurements may vary by period.
The primary result is a modeled dose-response estimate, but coefficient estimates, model diagnostics, goodness-of-fit information, and a formal test of the dose trend are not reported. The primary description also refers to adjustment by fluticasone furoate alone, while the posted dose-specific estimates are presented as expected responses; the record does not provide further detail on the full model specification or the exact contrast interpretation.
The number contributing to each primary estimate varies, but reasons for missing measurements and period-specific completeness are not reported. The record does not report imputation or sensitivity analyses. Without those details, the effect of incomplete measurements on precision or potential bias cannot be evaluated.
The registry lists multiple secondary outcomes and analyses but does not report a multiplicity plan. The secondary p-values are therefore not documented as adjusted for the number of outcomes or comparisons. The record also does not report a clinical-importance threshold for FEV1 or the secondary measures. Statistical estimates cannot by themselves determine whether the magnitude matters to participants.
Finally, the endpoint measurements cover short treatment periods, while the registry description gives approximately 14 weeks as the total participation duration. The posted results do not establish longer-term outcomes. These limits are specific to the design and information recorded and should temper conclusions drawn from the dose pattern, secondary findings, and safety counts.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Evaluate the Safety, Efficacy and Dose Response of GSK573719 in Combination With Fluticasone Furoate in Subjects With Asthma — NCT01573624.
- Registry record last updated 2017-10-06; status last verified 2017-09.
- Results first posted 2017-08-15.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01573624 was a completed Phase 2 randomized, double-blind, multicenter crossover study of treatments for asthma, sponsored by GlaxoSmithKline. The registry reports enrollment of 421 participants and describes three treatment periods, with each eligible participant assigned a sequence of three of seven potential treatments.
The primary endpoint was model-predicted change from baseline in trough FEV1, measured in litres and reported as a mean with a 95% confidence interval. The posted estimates increase across the five fluticasone furoate and UMEC dose combinations. Secondary outcomes cover morning and evening peak expiratory flow and rescue albuterol or salbutamol use.
The registry defines the primary analysis population as randomized participants who received at least one dose, with analyses limited to available data at the relevant time points. It does not report a multiplicity plan, detailed missing-data methods, or the crossover-specific model details needed to assess all analytical assumptions. Safety listings show no deaths and few serious adverse-event reports. The findings are short-term, endpoint-specific estimates and should be interpreted within those reporting limits.